Computer vision, user segment, and missing item determination
Techniques and systems are described that leverage computer vision as part of search to expand functionality of a computing device available to a user and increase operational computational efficiency as well as efficiency in user interaction. In a first example, user interaction with items of digital content is monitored. Computer vision techniques are used to identify digital images in the digital content, objects within the digital images, and characteristics of those objects. This information is used to assign a user to a user segment of a user population which is then used to control output of subsequent digital content to the user, e.g., recommendations, digital marketing content, and so forth.
1. A method comprising:
monitoring user interaction with a plurality of digital content, wherein the user interaction is manipulation by a user of digital images of the digital content via a user interface to focus on objects or characteristics of the objects in the digital images;
identifying the digital images within the plurality of digital content that are subject of the user interaction;
recognizing the objects included within the identified digital images or the characteristics of the recognized objects that are subject of the user interaction;
assigning a user corresponding to the user interaction to a user segment of a user population, the assigning including generating at least one machine learning model based on the recognized objects or the recognized characteristics;
generating a recommendation based on the assigned user segment; and
controlling output of a subsequent item of digital content based on the generated recommendation.
2. The method as described in claim 1 , wherein the manipulation corresponds to a gesture that focuses on the objects or that focuses on the characteristics of the objects.
3. The method as described in claim 1 , further comprising determining whether the user is interested in the objects or the characteristics of the objects.
4. The method as described in claim 3 , wherein determining whether the user is interested in the objects or the characteristics of the objects is based on whether the manipulation of the digital images results in the objects being recognizable using the machine learning model.
5. The method as described in claim 1 , wherein the monitoring includes identifying a first digital image included in digital content that is subject to the user interaction and a second digital image that is not, and wherein the generated recommendation is based on the identifying.
6. The method as described in claim 5 , wherein the digital content is configured as a webpage or screen of a user interface of an application.
7. The method as described in claim 1 , wherein controlling the output is based on the at least one machine learning model.
8. The method as described in claim 1 , wherein the manipulation indicates that the user is interested in the objects or characteristics of the objects and also indicates that the user is not interested in another object or characteristic of the other object in a digital image that is not manipulated.
9. A computing device comprising:
a processing system; and
a computer-readable storage medium having instructions stored thereon that, responsive to execution by the processing system, causes the processing system to perform operations comprising:
monitoring user interaction with a plurality of digital content, wherein the user interaction is a manipulation of the user interface to focus on objects or characteristics of the objects in digital images of the plurality of digital content;
identifying the digital images within the plurality of digital content that are subject of the user interaction;
recognizing the objects included within the identified digital images or the characteristics of the recognized objects that are subject of the user interaction;
assigning a user corresponding to the user interaction to a user segment of a user population, the assigning including generating at least one machine learning model based on the recognized objects or the recognized characteristics;
generating a recommendation based on the assigned user segment; and
controlling output of a subsequent item of digital content based on the generated recommendation.
10. The computing device as described in claim 9 , wherein the instructions further cause operations to be performed by the processing system including:
identifying digital images that are not subject to the user interaction based on the monitoring;
recognizing objects included within the identified digital images and characteristics of the recognized objects that are not subject of the user interaction; and
wherein the assigning is also based on the recognizing objects included within the identified digital images and characteristics of the recognized objects that are not subject of the user interaction.
11. The computing device as described in claim 9 , wherein the identifying includes identifying which digital images are subject to the user interaction and digital images that are not, and wherein the recognizing is performed for the digital images that are subject to the user interaction and is not performed for digital images that are not subject of the user interaction.
12. A non-transitory computer-readable storage medium comprising instructions, which when executed by one or more processors of a computing device, cause the computing device to perform operations comprising:
monitoring user interaction with a plurality of digital content, wherein the user interaction is manipulation by a user of digital images of the digital content via a user interface to focus on objects or characteristics of the objects in the digital images;
identifying the digital images within the plurality of digital content that are subject of the user interaction;
recognizing the objects included within the identified digital images or the characteristics of the recognized objects that are subject of the user interaction;
assigning a user corresponding to the user interaction to a user segment of a user population, the assigning including generating at least one machine learning model based on the recognized objects or the recognized characteristics;
generating a recommendation based on the assigned user segment; and
controlling output of a subsequent item of digital content based on the generated recommendation.
13. The computer-readable storage medium of claim 12 , wherein the operations further include identifying which digital images are subject to the user interaction and digital images that are not.
14. The computer-readable storage medium of claim 12 , wherein the plurality of digital content further includes webpages or screens of a user interface of an application executed by the computing device.
15. The computer-readable storage medium of claim 12 , wherein the recognizing is performed for the digital images that are subject to the user interaction and is not performed for digital images that are not subject of the user interaction.
16. The method as described in claim 1 , wherein the manipulation corresponds to a spoken utterance to focus on the objects or to focus on the characteristics of the objects.
17. The method as described in claim 1 , wherein the manipulation corresponds to zooming in or zooming out to focus on the objects or to focus on the characteristics of the objects.
18. The method as described in claim 1 , wherein the manipulation corresponds to a spoken utterance to perform a zooming in or perform a zooming out, the zooming in and the zooming out being performed to focus on the objects or to focus on the characteristics of the objects.
19. The computing device as described in claim 9 , wherein the manipulation corresponds to a gesture that focuses on the objects or that focuses on the characteristics of the objects.
20. The computer-readable storage medium of claim 12 , wherein the manipulation corresponds to a gesture that focuses on the objects or that focuses on the characteristics of the objects.